Human Reproduction Vol.20, No.12 pp. 3452–3458, 2005
doi:10.1093/humrep/dei241
Advance Access publication August 11, 2005.
Cyclic FEE peptide increases human gamete fusion
and potentiates its RGD-induced inhibition
A.Ziyyat1, N.Naud-Barriant1, V.Barraud-Lange1, F.Chevalier1, O.Kulski1, T.Lemkecher1,
M.Bomsel2 and J.P.Wolf1,3
1
Laboratoire de Biologie de la Reproduction, UPRES 3410, UFR SMBH, Université Paris 13, 74, rue Marcel Cachin, 93017 Bobigny,
Service d’Histologie-Embryologie-Cytogénétique, Hôpital Jean Verdier (Assistance Publique-Hôpitaux de Paris), Bondy and
2
Laboratoire ‘Entrée muqueuse du VIH et immunité muqueuse’, Unité INSERM U567, Institut Cochin, 22, rue Méchain, 75014 Paris,
France
3
To whom correspondence should be addressed. E-mail: [email protected]
BACKGROUND: 61 integrin has been proposed to act as a sperm receptor on the mouse oocyte by interacting
with spermatozoon fertilin . We investigated, in humans, whether oocyte integrins could act similarly in gamete
fusion, using a cyclic peptide containing the putative disintegrin-binding domain of human fertilin [cyclic FEE
(cFEE)] and RGD peptide. METHODS: Zona-free eggs were inseminated in the absence or presence of peptides. To
maintain the membrane protein pattern, the zona pellucida was removed by microdissection. Immunofluorescence
and confocal microscopy were used to detect integrin subunits on the oocyte. RESULTS: Unexpectedly, cFEE alone
increased human gamete fusion by 94% instead of inhibiting fertilization. Furthermore, cFEE together with RGD
potentiated the RGD-induced inhibition of fertilization in a dose-dependent manner. The data suggested the hypothesis of integrin cross-talk, further supported by the co-localization of 61 and v3 integrins, the putative receptors
of cFEE and RGD peptides, respectively. CONCLUSIONS: RGD-sensitive and -insensitive integrins may be associated in a multimolecular complex working as a sperm receptor on the human oocyte membrane. Supplementation of
human IVF culture medium with cFEE peptide might improve fertilization rates in ART.
Key words: FEE/fertilization/human/integrins/RGD
Introduction
The fusion process in all mammalian gametes is mediated by a
series of molecular interactions in which members of the tetraspanin, integrin and ADAM (A Disintegrin And Metalloprotease) families have been suggested to play a role (Wassarman
et al., 2001; Primakoff and Myles, 2002). The role of CD9
tetraspanin is now clearly established since homozygous null
females exhibit severely reduced fertility (Le Naour et al., 2000;
Miyado et al., 2000). In contrast, the role of oolemma integrins
is still debatable (Miller et al., 2000; He et al., 2003).
Integrins constitute a large family of heterodimeric transmembrane receptors composed of covalently linked α and β
subunits. The major function of these cell surface molecules is
to mediate cell–cell and cell–extracellular matrix attachment
(for a review see Bowen and Hunt, 2000). Integrin activation
induces a conformational change of their outer α and β
domains, inducing binding with ligands (Brakebusch and
Fassler, 2003). Ligand binding induces, in turn, integrin clustering at the cell surface, and recruitment of actin filaments and
signalling proteins to the cytoplasmic domain of integrins
(Giancotti and Ruoslahti, 1999; Hynes, 2002). Conversely,
intracellular signals can induce integrins to bind to their matrix
ligands (Liddington and Ginsberg, 2002). Eighteen α and eight
β subunits have been identified and form 23 known heterodimers (Zhu and Evans, 2002). Among the integrins, some mediate
preferentially adhesion to ligands such as fibronectin, vitronectin and fibrinogen by binding to the RGD motif present in the
molecule. They are known as ‘RGD-sensitive’. The RGD
motif has also been found in snake venom proteins that specifically inhibit integrin binding function and serve as potent
integrin antagonists. The majority of these proteins interact
with β1- and β3-associated integrins and their potency is at
least 500–2000 times higher than that of short RGD peptides
(Lu et al., 2003).
In reproductive biology, there is strong evidence that integrins
and their ligands are important mediators of sperm–egg binding. Both spermatozoa and oocytes express a number of
integrins and molecules that contain integrin recognition sites.
The mouse oocyte contains integrin subunits α2, α3, α5, α6,
αv, β1, β3 and β5 as detected by mRNA and/or protein analysis (Tarone et al., 1993; Evans et al., 1995b; Zuccotti et al.,
1998; Burns et al., 2002). Similarly, human oocytes express
integrin subunits α2, α3, α5, α6, αv, αM, β1, β2, β3, β4, β5
and β6 (Campbell et al., 1995; Ji et al. 1998; Sengoku et al.,
2004). Among the many different integrin subunits expressed
on oocytes, a large amount of data clarifies the major role of
3452 © The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
Cyclic FEE peptide increases human gamete fusion
α6β1 integrin in the process of gamete binding and/or fusion
(Almeida et al., 1995; Takahashi et al., 2000) though it is called
into question by others (Evans et al., 1997b; Evans, 1999; Miller
et al., 2000). However, a kinetic experiment revealed that, at an
early stage of sperm–egg fusion, the integrin α6 subunits accumulate at the sperm-binding site. The frequency of cluster
formation is closely related to that of sperm–egg fusion. This
led Takahashi et al. to suggest that sperm–egg fusion actually
occurs at sites where the α6β1 integrin is clustered and that this
interaction occurs via direct association of integrin α6 with
sperm (Takahashi et al., 2000).
Besides oocyte integrins, their putative ligands on sperm, i.e.
ADAMs such as fertilin β (ADAM2) and cyritestin (ADAM3),
have also been well studied as candidate genes responsible for
the sperm–oocyte interaction. Fertilin is an αβ heterodimer
first described on guinea pig spermatozoa as the putative ligand by which the sperm interacts with the oocyte (Primakoff
et al., 1987). Fertilin α and β subunits are the prototypes of the
ADAM gene family, which are a widely distributed family of
evolutionarily conserved transmembrane proteins (Blobel,
1992; Wolfsberg, 1995, 1996). The disintegrin domain of
ADAMs consists of 60–90 amino acids and is homologous to
snake venom disintegrins. The snake disintegrin domains contain RGD sequences and competitively inhibit integrin-mediated adhesion of platelets to RGD sequences in fibrinogen and
other ligands (Gould et al., 1990). Structural analysis has
revealed that the RGD sequence is present at the top of an
extended loop structure, similar to the location of the RGD
sequence in fibronectin (Leahy et al., 1996). However, most
ADAMs do not contain an RGD sequence in their disintegrin
domain. Indeed, the putative disintegrin tripeptide-binding site
of fertilin presents species specificity. In the position of the
RGD tripeptide, the consensus sequence in fertilin β (based on
cDNA clones) is QDE in the mouse and FEE in the human
(Gupta et al., 1996; for a review see Evans, 2002).
In addition to the data obtained in vitro, knockout mice for
these genes provided evidence for the participation of the
ADAMs in the sperm–oocyte interaction. Thus, fertilin β–/–
sperm show greatly reduced levels of binding to the oolemma,
but the few that do bind are still able to fuse (Cho et al., 1998).
Moreover, some experiments suggest that integrin α6β1 mediates the binding of fertilin β and cyritestin (Chen et al., 1999;
Yuan et al., 1999; Takahashi et al., 2001; Tomczuk et al.,
2003). Indeed, in the mouse, based on the capacity of specific
monoclonal antibodies (mAbs) to inhibit fertilin β binding to
the oocyte and the direct binding to this integrin of fertilin β
peptides, integrin α6β1 has been suggested to be a receptor for
fertilin β (Chen and Sampson, 1999; Chen et al., 1999; Bigler
et al., 2000).
On the other hand, members of the RGD-binding integrin
subfamily (including αvβ1, αvβ3 and αvβ5) are expressed by
oocytes and are implicated in fertilization by studies using
RGD peptides in IVF assays (Bronson and Fusi, 1990; Ji et al.,
1998) and other work (Linfor and Berger, 2000). Bronson and
Fusi (1990) demonstrated that RGD-dependent recognition is
involved in sperm–oolemmal adhesion since RGD peptides
inhibit the fertilization rate in the zona-free hamster egg penetration test. Furthermore, the oligopeptide GdRGDSP, specifically
designed to block both fibronectin and vitronectin receptors,
significantly inhibits the binding of human sperm to the oolemma
of zona pellucida-free hamster oocytes (Fusi et al., 1996). This
is of importance since fibronectin and vitronectin are two glycoproteins that contain functional RGD sequences and are both
present on human spermatozoa (Fusi et al., 1992, 1994).
Integrins that recognize these ligands have been detected on
spermatozoa and eggs. We similarly reported a dose-dependent
and partial inhibition of the human gamete fusion by RGD
peptide (Ji et al., 1998). These results indicate that an RGDsensitive integrin is involved in the fusion process. Biochemical analyses have implicated the αv integrin subunit on the pig
oocyte in the recognition of isolated pig sperm membrane proteins (Linfor and Berger, 2000), but direct involvement of
RGD-sensitive integrins should be confirmed.
Despite this supportive evidence, a precise role for ADAMs
and integrins, RGD sensitive or not, remains to be determined.
Moreover, these data will also be discussed with regard to
experiments showing that (i) antibody inhibition studies using
different conditions did not find inhibition of sperm–egg
fusion by an anti-α6 mAb (GoH3) (Evans et al., 1997b;
Evans, 1999; Miller et al., 2000); (ii) oocytes from α6 knockout mice showed normal binding and fusibility with sperm
(Miller et al., 2000); and (iii) using β1 conditionally deficient
females, α3 integrin null mice and anti-β3 or αv integrin function-blocking antibodies, α3β1 is not essential for sperm–egg
binding and fusion and β1 integrin null eggs are fully functional in fertilization both in vivo and in vitro (He et al., 2003).
These results could not totally exclude the possibility of the
occurrence of functional redundancy among various integrins
and/or of a macromolecular complex including integrins and
tetraspanins.
In humans, the fertilization mechanism is rarely studied,
mainly because oocytes are rare and precious. Similarly to
mouse models, ADAMs and several integrins, including the α6
and β1 subunit, have been suggested to mediate sperm–egg
interaction (Fusi et al., 1992; Campbell et al., 1995; Ji et al.,
1998) but, once again, these data are called into question by
other work (Sengoku et al., 2004). To address this conundrum,
we studied the human gamete interaction process using peptide
mimicking the putative binding site of the disintegrin domain
of human fertilin β. Effectively, since fertilin α and cyritestin
genes are non-functional in human (Jury et al., 1997; Grzmil
et al., 2001), fertilin β appears to be the best candidate for
involvement in gamete interaction. As the disintegrin-binding
domain of fertilin β is localized at the top of a hairpin loop, we
chose to use a cyclic hexapeptide (CSFEEC). Moreover,
because the RGD peptide, which inhibits sperm–egg fusion,
does not bind to α6β1 integrin, we co-incubated the two peptides. We also propose a model of a membrane receptor in the
human gamete fusion process which includes integrins and the
CD9 tetraspanin.
Materials and methods
Human gamete preparation
Immature and unfertilized human oocytes were donated by patients
undergoing ICSI and IVF, respectively. Exceptionally, a fresh cohort
3453
A.Ziyyat et al.
of mature oocytes was given when no spermatozoa were available.
Similar results were obtained with these fresh oocytes to those with
unfertilized or in vitro matured oocytes for immunostaining and sperm–
egg fusion assay. Informed consent was obtained from all patients and
this study was approved by the local ethical committee.
For IVF, normal metaphase II (MII) stage human oocytes were
inseminated and examined for the presence of pronuclei 18 h postinsemination. At day 2 post-insemination, oocytes with no signs of
fertilization and apparently normal morphology were donated and
included in this study.
Following oocyte collection, oocytes intended for ICSI were freed
of the surrounding cumulus and corona radiata cells by brief exposure
to hyaluronidase (type VIII, 80 IU/ml; Sigma) at 37°C, and gentle
pipetting with a Stripper® micropipettor (135 μm diameter; MidAtlantic Diagnostics, Marlton, USA). They were then scored for the
presence of a germinal vesicle (GV) or a polar body. In vitro matured
metaphase II oocytes were obtained by incubating GV or metaphase I
stage (neither GV nor polar body was present) oocytes under a 5%
CO2 atmosphere at 37°C, in FertiCult™ medium (FertiPro, Beemem,
Belgium) for 24 h. Zona pellucida were removed mechanically with a
totally chemical and enzymatically free technique, using a pair of
microdissection scissors under light Stero-microscope, in order to
preserve the integrity of the membrane’s protein pattern (Kellom
et al., 1992).
Semen from donors of proven fertility was collected after 3 days of
sexual abstinence. Sperm samples were kept at 37°C until liquefaction
was completed. A two-step 90/45% PureSperm® gradient (Nidacon
International, Gothenburg, Sweden) was used to select motile spermatozoa (300 g for 20 min). The pellet was washed once with FertiCult™ medium by centrifugation and resuspended to a sperm
concentration of 2 × 106/ml for the experiments. Semen analysis was
performed according to World Health Organization criteria (World
Health Organization, 1999). The sperm were kept under capacitating
conditions for 2 h before insemination (Mortimer et al., 1989).
Antibodies and peptides
MAb against the human αvβ3 integrin (LM609) and rat mAb against
the α6 integrin subunit (GoH3) were purchased from Chemicon International (London, UK). Function blocking mAb against the human β1
integrin subunit (DE9) (Bergelson, 1992) was purchased from UBI
(New York, USA). Donkey anti-mouse rhodamine conjugate and a
rabbit anti-rat–fluorescein isothiocyanate (FITC) conjugate, mouse
adsorbed (Vector Laboratories, Burlingame, CA), were used.
The sequence of the disintegrin domain of human ADAM2 is
CLFMSKERMCRPSFEECDLPEYCNGSSASC
(accession
no.
CAA67753). The FEE peptide was synthesized by Neosystem (Strasbourg, France). Its formula, CSFEEC, contained the tripeptide FEE
and was cyclized by the adjunction of a cysteine at both ends. The
peptide was purified by high-pressure liquid chromatography to >95%
purity. The scrambled cyclic peptide CFESEC was obtained in a similar way. For immunofluorescence detection, a biotinylated and
cyclized FEE peptide was obtained by adjunction of a biotinylated
Gly–Gly chain. RGD-containing peptide (Gly-Arg-Gly-Asp-Thr-Pro)
was purchased from Sigma.
Fluorescence staining of human zona-free eggs
Single or double staining were performed by incubating zona-free
eggs with primary antibodies for 1 h at the following concentrations:
anti-α6 (20 μg/ml), anti-β1 (10 μg/ml) and anti-αvβ3 (10 μg/ml).
These incubations were followed by washing with FertiCult™
medium prior to the staining procedure with rabbit anti-rat–FITC
conjugate, mouse adsorbed (10 μg/ml), and/or rhodamine-conjugated
3454
donkey anti-mouse antibody (20 μg/ml). Eggs were then fixed in 2%
paraformaldehyde in phosphate-buffered saline (PBS) at room
temperature for 30 min. Oocytes were mounted in Immu-mount antifade solution (Shandon, Pittsburgh, PA), and confocal analysis was
performed with a TCS SP2 confocal microscope (Leica, Wetzlar,
Germany), using a 63× objective. Negative controls were obtained
by substituting the incubation in primary antibody by incubation in
PBS–bovine serum albumin (BSA) 1% alone. Non-immune rat
immumnoglobulin IgG 2a was also used as control. To verify that
clustering was not mediated by the antibodies, labellings were also
performed after paraformaldehyde fixation and identical results
were obtained.
To detect FEE binding on human eggs, zona-free oocytes were
incubated for 1 h with biotinylated cyclic FEE (cFEE; 100 μmol/l).
Then, they were incubated with an anti-biotin mAb (Zymed Laboratories, San Francisco, CA), a biotin-conjugated anti-mouse IgG (Chemicon International) and finally with streptavidin–FITC. Detection was
performed using confocal microscopy (Bio-Rad MCR1000) on a
Nikon Diaphot 300.
Human sperm–egg fusion assay
Due to French bioethical laws, in vitro human gamete fusion assays
can only be performed using zona pellucida-free oocytes. Zona pellucida-free mature eggs were pre-incubated with different concentrations of peptides for 30 min in 20 μl FertiCult™ medium drops
and then inseminated with 4000 capacitated motile human spermatozoa, in the presence of the same peptides, under mineral oil. Controls were obtained by omitting the peptide or using a scrambled
peptide (CFESEC). After 18ˆh, oocytes were washed and loaded
with DNA-specific fluorochrome Hoechst 33342 (Sigma) at 5 μg/ml
for 20 min. After washing, they were fixed in 4% paraformaldehyde
in PBS–BSA 1% for 30 min at room temperature. Eggs were
mounted in slides and analysed using a Zeiss Axiophot microscope
equipped with a camera and connected to an Imaging System Package (Applied Imaging, Newcastle-upon-Tyne, UK). Spermatozoa
were considered as fused when their nucleus became Hoechststained and decondensed.
Statistical analysis
Statistical analysis of the data was determined by Statview® package.
Means were compared by non-parametric Wilcoxon test. Differences
were considered significant at P < 0.05.
Results
The 61 and v3 integrins are present on human oocyte
membrane
The immunofluorescence study showed the presence of the α6
and β1 integrin subunits at the surface of the human zona-free
egg. They both formed patches (Figure 1A, B and D). Merging
of the α6 and β1 subunit staining (Figure 1C) showed a colocalization of both integrin subunits in patches, suggesting the
existence of oolemma α6β1 integrin dimers. Patches of αvβ3
integrin were also found at the human zona-free oocyte surface
(Figure 1E). Merging of α6 and αvβ3 patches showed a colocalization of these molecules (Figure 1F). These patches
were observed whether fixation was performed after or before
labelling and were similar whether labelling was done at 4°C
or at room temperature, indicating that they were not induced
by the mAb after zona pellucida removal.
Cyclic FEE peptide increases human gamete fusion
gametes were incubated with 100 μmol/l cFEE. On the contrary, we observed an increase in the number of fused spermatozoa per oocyte (36.9 ± 11.7, P < 0.0001) (Figure 3B and
E). Scrambled peptide had no effect on gamete fusion (21.4 ± 4.3,
NS) (Figure 3E).
Cyclic FEE peptide enhances the RGD-induced inhibition of
human gamete fusion
Figure 1. Localization by confocal microscopy of integrin subunits
α6 and β1, and αvβ3 integrin on zona pellucida-free human oocytes.
Zona-free eggs were labelled for integrin subunits α6 (A and D) or β1
(B) or integrin αvβ3 (E) by indirect immunostaining and confocal
detection. Images correspond to projections of consecutive optical
sections of half oocytes. (C) is the merged image of (A) and (B), and
(F) is the merged image of (D) and (E). The co-localization of patches
suggests the presence of α6β1 dimers on the human oocyte membrane
and its co-localization in a multimolecular receptor with αvβ3
integrin.
Figure 2. Binding of the cyclic FEE peptide to human oocytes.
Zona-free human eggs were incubated with biotinylated cFEE peptide. Bound peptide was detected by indirect immunofluorescence.
Observation was performed using confocal microscopy. (A) An equatorial section. (B) Superposition of consecutive sections of a half
oocyte. The scrambled peptide, CFESEC, did not bind to the egg.
Cyclic FEE peptide binds to the human oocyte membrane
and increases the number of fused spermatozoa
Immunofluorescence study using the biotinylated cFEE, but
not the scrambled peptide, and confocal microscopy showed
that it binds to the oolemma of human zona-free oocyte and
forms patches (Figure 2A and B). We therefore decided to
perform a sperm–egg fusion assay using the cFEE peptide.
According to previously reported experiments in a mouse
model (Zhu and Evans, 2002) or in the zona-free hamster
oocyte penetration test (Bronson et al., 1999) which showed
an inhibition of fertilization with a linear peptide, the same
result was expected in a homogeneous human system. Surprisingly, while in control oocytes a mean of 19.0 ± 4.6
(mean ± SD) spermatozoa were found in the cytoplasm (Figure 3A and E), there was no inhibition of fertilization when
As RGD peptide was known to be an inhibitor of gamete
fusion, we tested the co-incubation of both cFEE and RGD
peptide in the insemination medium. Co-incubation of 100 μmol/l
cFEE and 100 μmol/l RGD peptide accentuated the inhibitory
effect of RGD. Indeed, only 4.7 ± 3.9 fused spermatozoa per
oocyte (Figure 3C and E) were observed versus 11.1 ± 3.2
(P < 0.0008) when fertilization occurs with RGD alone (Ji
et al., 1998). Increasing the concentration of RGD to
200 μmol/l induced a dose-dependent decrease in the number
of fused spermatozoa with a mean ± SD of 1.7 ± 2.2 (P < 0.0001)
(Figure 3E). Some of these oocytes presented a complete
fusion failure and only the egg metaphase could be seen
(Figure 3D).
Discussion
The cFEE peptide mimicking the ADAM2 disintegrin-binding
domain increases the human oocyte fertilization index. The same
effect was obtained in mouse with an equivalent peptide (data
not shown). This was a surprise since comparable experiments
produced the opposite results in mouse and human. Indeed,
recombinant fertilin β and a peptide corresponding to the fertilin
β disintegrin loop inhibit sperm oocyte binding and reduce the
incidence of fertilization in the mouse (Almeida et al., 1995;
Evans et al., 1997b; Bigler et al., 2000; Gupta et al., 2000; Zhu
et al., 2000) and in human (Bronson et al., 1999). Peptides used
in mouse studies were either recombinant (Evans et al., 1997b)
or synthesized, e.g. CAQDEC (Evans et al., 1995a) or AQDECDVT (Zhu et al., 2000). They all contained the tripeptide
QDE and some other adjacent amino acids. The peptide used in
the human study was SFEECDLP (Bronson et al., 1999).
Both the composition and the cyclization of the peptide can
explain the absence of an inhibitory effect. It lacked the second
aspartic acid of the predicted disintegrin loop pentapeptide
sequence QDECD, which has been shown to be critical for its
inhibitory properties (Zhu et al., 2000). Our peptide did not
contain this terminal aspartate which seems to be responsible
for the inhibitory effect. Furthermore, our peptide was cyclized.
It has been shown that restricting the conformational space of
active peptide sequences, by using them in cyclic form, could
lead to components with improved activity and receptor selectivity. This has been demonstrated for a synthetic cyclic RGDcontaining peptide which was a 20- to 100-fold better inhibitor
of cell adhesion to vitronectin and/or laminin fragments when
compared with a linear variant (Pierschbacher and Ruoslahti,
1987; Aumailley et al., 1991). Such hypothesis could explain
our peptide’s activity. Similarly, linear CAQDEC has been
shown to reduce sperm binding and sperm–egg fusion, while
the cyclic form did not have any inhibitory effect (Evans et al.,
1995a). This result may, however, be species dependent since
3455
A.Ziyyat et al.
Figure 3. Effects of the cFEE and RGD-containing peptides on the fertilization of zona-free human oocytes. Zona-free human eggs were inseminated with human spermatozoa in the absence (A and E) or in the presence of cFEE peptide at 100 μmol/l (B and E), alone or associated with
RGD-containing peptide at 100 μmol/l (C and E) or 200 μmol/l (D and E). Fused spermatozoa were counted under UV excitation. Spermatozoa
were considered as fused when their nuclei were Hoechst 33342-stained and decondensed. (E) Means ± SD of different experiments. Numbers
above represent the number of human oocytes in each group. *Significantly different from the control (P < 0.0001); **significantly different from
RGD alone (P < 0.0008); ***significantly different from FEERGD1000 (P < 0.0001).
Myles et al. (1994) found that an equivalent cyclic peptide was
an inhibitor in the guinea pig.
There is, in fact, considerable variation and inconsistency in
the literature regarding the efficacy of peptide mimics used in
such IVF assays. The technique of oocyte depellucidation may
be critical since enzymatic techniques may modify their membrane fertilin receptors (Boldt et al., 1988, 1989). For example,
different zona removal techniques induce different distributions of the α6 integrin subunit (Tarone et al., 1993; Zuccotti
et al., 1998; Miller et al., 2000; Takahashi et al., 2000). To prevent any protease digestion of membrane proteins and eventually of the receptor we are interested in, we performed a totally
mechanical zona pellucida removal technique using a pair of
microdissection scissors under Stero-microscope without any
enzyme or chemical.
As already shown, we found a partial inhibitory effect of
RGD peptide on gamete fusion (Ji et al., 1998). Interestingly,
it is potentiated by the cFEE. Hence, in the presence of
cFEE, the dose–response effect of RGD achieved an almost
complete inhibition for 200 μmol/l. Bronson et al. (1999)
described a similar potentiation of the inhibitory effect of RGD
peptide by FEE. They used a linear FEE peptide and reported
that combined SFEECDLP and G4120, a cyclic RGD-containing peptide, exhibited a strong inhibition of both adhesion and
penetration at concentrations that individually had been ineffective, suggesting cooperation between the two receptor ligands during fertilization in human. They suggested that two
different integrins should be involved in human gamete interaction, one of which recognizes fertilin and one which recognizes RGD-containing sperm-associated proteins such as
vitronectin, and that they may cooperate (Bronson et al., 1999).
Actually, we show that the α6β1 integrin is localized on the
surface of human oocyte in patches. Interestingly, RGD-sensitive
αvβ3 integrin is also present on the oocyte membrane and
co-localizes with the α6β1 integrin, suggesting their direct
cooperation within a functional complex (Figure 1F). In the
mouse, Takahashi et al. (2000) also showed that α6 patches
appear on the oocyte membrane at the sperm-binding site prior
to fusion. We thus suggest that the functional complex to
3456
which the fertilizing sperm binds contains at least these two
RGD-sensitive and RGD-independent integrins.
On the other hand, the tetraspanin CD9 is essential for sperm–
egg fusion since female mice that have been deleted for this gene
are infertile. Their infertility is related to the inability of the
oolemma to fuse with the sperm (Le Naour et al., 2000; Miyado
et al., 2000). Tetraspanins are surface proteins containing four
transmembrane domains and forming complexes with each other
and with various membrane proteins within a network of molecular interactions called the ‘tetraspanin web’ (Berditchevski,
2001; Boucheix and Rubinstein, 2001; Hemler, 2003). They
contribute to the formation of cell surface multimolecular complexes and thereby may participate in the functional regulation
of molecules with which they associate (Berditchevski, 2001;
Boucheix and Rubinstein, 2001; Hemler, 2003). One hypothesis
is that CD9 may be necessary for the patch formation. This could
explain the inability of CD9-deleted oocytes to fuse since they
are unable to form α6 integrin patches (data not shown).
Finally, the oocyte sperm receptor could be composed of at
least two integrins linked by tetraspans, α6β1 and αvβ3, RGDinsensitive and RGD-dependent, respectively. Effectively, both
α6β1 and β3 are linked to CD9 in somatic cells. α6β1 integrin is
linked to CD9 via CD151 (Serru et al., 1999) and the β3 integrin
subunit is associated with CD9 in platelets (Longhurst et al.,
1999). One can therefore speculate that the fusing spermatozoon
first binds loosely to α6β1 and induces α6β1 and αvβ3 patch
formation. It can then bind tightly to the oocyte through the
αvβ3 integrin prior fusion (Figure 4). Fusi et al. (1996) actually
have described the vitronectin receptor as being the possible
‘velcro’ whereby the sperm binds to the oocyte. αvβ3 is, precisely, the vitronectin receptor (Fusi et al., 1996). The fact that
we detected its presence on the human oocyte surface and its
co-localization with the α6β1 integrin supports this hypothesis.
This model allows us to propose the following interpretation
of the reported data. When the two peptides are co-incubated,
the cFEE peptide induces the recruitment and/or the activation
of all available αvβ3 integrin-binding sites on which the RGD
peptide binds. Increasing the RGD concentration results in
fewer and fewer available binding sites. At 200ˆμmol/l, there
Cyclic FEE peptide increases human gamete fusion
(Nishimura et al., 2001; He et al., 2003; for review see Stein
et al., 2004). Finally, redundant molecules may take the place of
the deleted protein and assume its functions.
In conclusion, it is the first time that a peptide with the capacity of increasing human gamete fertilizing ability is reported.
Such molecule could be used for supplementing the insemination medium to improve the fertilization rate in human IVF and
perhaps limit the application of ICSI in assisted reproduction.
Despite contradictory data from invalidated gene models, we
believe that, physiologically, integrins participate in a multimolecular complex with tetraspanin and probably other as yet
undiscovered molecules.
Acknowledgements
Figure 4. Model of human gamete membrane interaction. Before
interaction (A). The binding of sperm fertilin β to oocyte α6β1
integrin (B) triggers multimolecular patch formation containing αvβ3
integrin via a tetraspanin web (C).
are absolutely no free αvβ3-binding sites where the sperm
could bind and fuse. This could explain the RGD dose-dependent inhibition of the gamete fusion process in the presence of
cFEE. On the contrary, when RGD peptide is incubated alone,
it binds to the available αvβ3 integrin. However, one can speculate that the fertilizing spermatozoon can find a free α6β1
integrin site to interact with and induce formation of a complex
containing αvβ3 integrin on which the sperm can bind competitively with RGD peptide. This could explain why RGDinduced inhibition of gamete fusion is always partial. When
cFEE peptide is incubated alone, it may induce numerous complex binding sites on which the inseminated spermatozoa can
bind and fuse easily, explaining the increase in the number of
fused sperm. This model also explains how the lack of tetraspanin CD9 in CD9–/– eggs prevents formation of integrin
patches and consequently gamete fusion.
Our findings are in contrast to previous reports stating that
integrins are not essential for sperm–egg fusion (He et al., 2003).
However, this assumption is debatable for several reasons. First,
there may be several ways in which the oocyte may fuse, some
being as yet unknown (Stein et al., 2004). However, in the physiological process, integrins are effectively present on the
oolemma and obviously concerned with the fusion process
according to previous reports (Fusi et al., 1992; Almeida et al.,
1995; Evans et al., 1997a; Takahashi et al., 2000). Secondly, the
removal of the zona pellucida may bypass a step of the fusion
process rendering, for example, the α6β1 integrin useless. This
would explain the fusing ability of α6 or β1 integrin genedeleted oocytes. Thirdly, experiments with deleted genes are difficult to interpret. Indeed, studies of knockout mice have shown
that frequently there are unanticipated associated defects. Hence
sperm from fertilin β and cyristestin knockout mice show
reduced binding to the oocyte plasma membrane as expected,
but they also adhere poorly to the zona pellucida and show deficient migration from the uterus to the oviduct (Cho et al., 1998).
Analysis of the protein profile of oocytes or sperm of these
knockout mice also shows multiple molecular deficiency
We thank P.Fontanges (Hôpital Tenon), M.Garfa (Institut Cochin)
and S.Chambris (Université Paris 13) for their technical assistance,
and H.Giraud and F.Wolf for imaging assitance. We also thank
B.Ducot for her help in statistical analysis, and Dr A.Hazout and
P.Cohen-Bacrie from Eylau laboratories for providing us with human
materials. This work was supported by UPRES 3410 from the French
Ministry of Research.
References
Almeida EA, Huovila AP, Sutherland AE, Stephens LE, Calarco PG, Shaw LM,
Mercurio AM, Sonnenberg A, Primakoff P, Myles DG et al. (1995) Mouse egg
integrin alpha 6 beta 1 functions as a sperm receptor. Cell 81,1095–1104.
Aumailley M., Gurrath M., Muller G., Calvete J., Timpl R. and Kessler H
(1991) Arg–Gly–Asp constrained within cyclic pentapeptides. Strong and
selective inhibitors of cell adhesion to vitronectin and laminin fragment P1.
FEBS Lett 291,50–54.
Berditchevski F (2001) Complexes of tetraspanins with integrins: more than
meets the eye. J Cell Sci 114,4143–4151.
Bigler D, Takahashi Y, Chen MS, Almeida EA, Osbourne L and White JM
(2000) Sequence-specific interaction between the disintegrin domain of
mouse ADAM 2 (fertilin beta) and murine eggs. Role of the alpha(6)
integrin subunit. J Biol Chem 275,11576–11584.
Blobel CP, Wolfsberg TG, Turck CW, Myles DG, Primakoff P and White JM
(1992) A potential fusion peptide and an integrin ligand domain in a protein
active in sperm–egg fusion. Nature 356,248–252.
Boldt J, Howe AM and Preble J (1988) Enzymatic alteration of the ability
of mouse egg plasma membrane to interact with sperm. Biol Reprod
39,19–27.
Boldt J, Gunter LE and Howe AM (1989) Characterization of cell surface
polypeptides of unfertilized, fertilized, and protease-treated zona-free mouse
eggs. Gamete Res 23,91–101.
Boucheix C and Rubinstein E (2001) Tetraspanins. Cell Mol Life Sci 58,1189–1205.
Bowen JA and Hunt JS (2000) The role of integrins in reproduction. Proc Soc
Exp Biol Med 223,331–343.
Brakebusch C and Fassler R (2003) The integrin–actin connection, an eternal
love affair. EMBO J 22,2324–2333.
Bronson RA and Fusi F (1990) Evidence that an Arg–Gly–Asp adhesion
sequence plays a role in mammalian fertilization. Biol Reprod 43,1019–1025.
Bronson RA, Fusi FM, Calzi F, Doldi N and Ferrari A (1999) Evidence that a
functional fertilin-like ADAM plays a role in human sperm–oolemmal interactions. Mol Hum Reprod 5,433–440.
Burns KH, Owens GE, Fernandez JM, Nilson JH and Matzuk MM (2002)
Characterization of integrin expression in the mouse ovary. Biol Reprod
67,743–751.
Campbell S, Swann HR, Seif MW, Kimber SJ and Aplin JD (1995) Cell adhesion molecules on the oocyte and preimplantation human embryo. Hum
Reprod 10,1571–1578.
Chen H and Sampson NS (1999) Mediation of sperm–egg fusion: evidence
that mouse egg alpha6 beta1 integrin is the receptor for sperm fertilin beta.
Chem Biol 6,1–10.
Chen MS, Almeida EA, Huovila AP, Takahashi Y, Shaw LM, Mercurio AM
and White JM (1999) Evidence that distinct states of the integrin alpha6
beta1 interact with laminin and an ADAM. J Cell Biol 144,549–561.
3457
A.Ziyyat et al.
Cho C, Bunch DO, Faure JE, Goulding EH, Eddy EM, Primakoff P and Myles
DG (1998) Fertilization defects in sperm from mice lacking fertilin beta.
Science 281,1857–1859.
Evans JP (1999) Sperm disintegrins, egg integrins, and other cell adhesion
molecules of mammalian gamete plasma membrane interactions. Front Biosci 4, D114–D131.
Evans JP (2002) The molecular basis of sperm–oocyte membrane interactions
during mammalian fertilization. Hum Reprod Update 8,297–311.
Evans JP, Schultz RM and Kopf GS (1995a) Mouse sperm–egg plasma membrane interactions: analysis of roles of egg integrins and the mouse sperm
homologue of PH-30 (fertilin) beta. J Cell Sci 108,3267–3278.
Evans JP, Schultz RM and Kopf GS (1995b) Identification and localization of
integrin subunits in oocytes and eggs of the mouse. Mol Reprod Dev
40,211–220.
Evans JP, Schultz RM and Kopf GS (1997a) Characterization of the binding of
recombinant mouse sperm fertilin alpha subunit to mouse eggs: evidence for
function as a cell adhesion molecule in sperm–egg binding. Dev Biol
187,94–106.
Evans JP, Kopf GS and Schultz RM (1997b) Characterization of the binding of
recombinant mouse sperm fertilin beta subunit to mouse eggs: evidence for
adhesive activity via an egg beta1 integrin-mediated interaction. Dev Biol
187,79–93.
Fusi FM, Vignali M, Busacca M and Bronson RA (1992) Evidence for the
presence of an integrin cell adhesion receptor on the oolemma of unfertilized human oocytes. Mol Reprod Dev 31,215–222.
Fusi FM, Lorenzetti I, Mangili F, Herr JC, Freemerman AJ, Gailit J and
Bronson RA (1994) Vitronectin is an intrinsic protein of human spermatozoa released during the acrosome reaction. Mol Reprod Dev 39,337–343.
Fusi FM, Bernocchi N, Ferrari A and Bronson RA (1996) Is vitronectin the
velcro that binds the gametes together? Mol Hum Reprod 2,859–866.
Giancotti FG and Ruoslahti E (1999) Integrin signaling. Science 285,1028–1032.
Gould RJ, Polokoff MA, Friedman PA, Huang TF, Holt JC, Cook JJ and
Niewiarowski S (1990) Disintegrins: a family of integrin inhibitory proteins
from viper venoms. Proc Soc Exp Biol Med 195,168–171.
Grzmil P, Kim Y, Shamsadin R, Neesen J, Adham IM, Heinlein UA,
Schwarzer UJ and Engel W (2001) Human cyritestin genes (CYRN1 and
CYRN2) are non-functional. Biochem J 357,551–556.
Gupta S, Li H and Sampson NS (2000) Characterization of fertilin beta–disintegrin binding specificity in sperm–egg adhesion. Bioorg Med Chem
8,723–729.
Gupta SK, Alves K, Palladino LO, Mark GE and Hollis GF (1996) Molecular
cloning of the human fertilin beta subunit. Biochem Biophys Res Commun
224,318–326.
He ZY, Brakebusch C, Fassler R, Kreidberg JA, Primakoff P and Myles DG
(2003) None of the integrins known to be present on the mouse egg or to be
ADAM receptors are essential for sperm–egg binding and fusion. Dev Biol
254,226–237.
Hemler ME (2003) Tetraspanin proteins mediate cellular penetration, invasion,
and fusion events and define a novel type of membrane microdomain. Annu
Rev Cell Dev Biol 19,397–422.
Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell
110,673–687.
Ji YZ, Wolf JP, Jouannet P and Bomsel M (1998) Human gamete fusion can
bypass beta1 integrin requirement. Hum Reprod 13,682–689.
Jury JA, Frayne J and Hall L (1997) The human fertilin alpha gene is non-functional: implications for its proposed role in fertilization. Biochem J 321,577–581.
Kellom T, Vick A and Boldt J (1992) Recovery of penetration ability in protease-treated zona-free mouse eggs occurs coincident with recovery of a cell
surface 94 kD protein. Mol Reprod Dev 33,46–52.
Leahy DJ, Aukhil I and Erickson HP (1996) 2.0 Å crystal structure of a fourdomain segment of human fibronectin encompassing the RGD loop and synergy region. Cell 84,155–164.
Le Naour F, Rubinstein E, Jasmin C, Prenant M and Boucheix C (2000) Severely
reduced female fertility in CD9-deficient mice. Science 287,319–321.
Liddington RC and Ginsberg MH (2002) Integrin activation takes shape. J Cell
Biol 158,833–839.
Linfor J and Berger T (2000) Potential role of alphav and beta1 integrins as oocyte
adhesion molecules during fertilization in pigs. J Reprod Fertil 120,65–72.
Longhurst CM, White MM, Wilkinson DA and Jennings LK (1999) A CD9,
alphaIIbbeta3, integrin-associated protein, and GPIb/V/IX complex on the
surface of human platelets is influenced by alphaIIb beta3 conformational
states. Eur J Biochem 263,104–111.
Lu X, Lu D, Scully MF and Kakkar VV (2003) Modulation of integrin-binding
selectivity by mutation within the RGD-loop of snake venom proteins: a
3458
novel drug development approach. Curr Med Chem Cardiovasc Hematol
Agents 1,189–196.
Miller BJ, Georges-Labouesse E, Primakoff P and Myles DG (2000) Normal
fertilization occurs with eggs lacking the integrin alpha6beta1 and is CD9dependent. J Cell Biol 149,1289–1296.
Miyado K, Yamada G, Yamada S, Hasuwa H, Nakamura Y, Ryu F, Suzuki K,
Kosai K, Inoue K, Ogura A et al. (2000) Requirement of CD9 on the egg
plasma membrane for fertilization. Science 287,321–324.
Myles DG, Kimmel LH, Blobel CP, White JM and Primakoff P (1994) Identification of a binding site in the disintegrin domain of fertilin required for
sperm–egg fusion. Proc Natl Acad Sci USA 91,4195–4198.
Nishimura H, Cho C, Branciforte DR, Myles DG and Primakoff P (2001)
Analysis of loss of adhesive function in sperm lacking cyritestin or fertilin
beta. Dev Biol 233,204–213.
Pierschbacher MD and Ruoslahti E (1987) Influence of stereochemistry of the
sequence Arg–Gly–Asp–Xaa on binding specificity in cell adhesion. J Biol
Chem 262,17294–17298.
Primakoff P and Myles DG (2002) Penetration, adhesion, and fusion in mammalian sperm–egg interaction. Science 296,2183–2185.
Primakoff P, Hyatt H and Tredick-Kline J (1987) Identification and purification of a sperm surface protein with a potential role in sperm–egg membrane
fusion. J Cell Biol 104,141–149.
Sengoku K, Takuma N, Miyamoto T, Horikawa M and Ishikawa M (2004)
Integrins are not involved in the process of human sperm–oolemmal fusion.
Hum Reprod 19,639–644.
Serru V, Le Naour F, Billard M, Azorsa DO, Lanza F, Boucheix C and Rubinstein E (1999) Selective tetraspan–integrin complexes (CD81/alpha4beta1,
CD151/alpha3beta1, CD151/alpha6beta1) under conditions disrupting
tetraspan interactions. Biochem J 340,103–111.
Stein KK, Primakoff P and Myles D (2004) Sperm–egg fusion events at the
plasma membrane. J Cell Sci 117,6269–6274.
Takahashi Y, Yamakawa N, Matsumoto K, Toyoda Y, Furukawa K and Sato E
(2000) Analysis of the role of egg integrins in sperm–egg binding and
fusion. Mol Reprod Dev 56,412–423.
Takahashi Y, Bigler D, Ito Y and White JM (2001) Sequence-specific interaction between the disintegrin domain of mouse ADAM 3 and murine eggs:
role of beta1 integrin-associated proteins CD9, CD81, and CD98. Mol Biol
Cell 12,809–820.
Tarone G, Russo MA, Hirsch E, Odorisio T, Altruda F, Silengo L and Siracusa
G (1993) Expression of beta 1 integrin complexes on the surface of unfertilized mouse oocyte. Development 117,1369–1375.
Tomczuk M, Takahashi Y, Huang J, Murase S, Mistretta M, Klaffky E, Sutherland A, Bolling L, Coonrod S, Marcinkiewicz C et al. (2003) Role of multiple beta1 integrins in cell adhesion to the disintegrin domains of ADAMs 2
and 3. Exp Cell Res 290,68–81.
Wassarman PM, Jovine L and Litscher ES (2001) A profile of fertilization in
mammals. Nat Cell Biol 3, E59–E64.
World Health Organization (1999) WHO Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interaction, 4th edn.
Cambridge University Press, Cambridge.
Wolfsberg TG and White JM (1996) ADAMs in fertilization and development.
Dev Biol 180,389–401.
Wolfsberg TG, Primakoff P, Myles DG and White JM (1995) ADAM, a novel
family of membrane proteins containing A Disintegrin And Metalloprotease
domain: multipotential functions in cell–cell and cell–matrix interactions. J
Cell Biol 131,275–278.
Yuan R, Primakoff P and Myles DG (1997) A role for the disintegrin domain
of cyritestin, a sperm surface protein belonging to the ADAM family, in
mouse sperm–egg plasma membrane adhesion and fusion. J Cell Biol
137,105–112.
Zhu X and Evans JP (2002) Analysis of the roles of RGD-binding integrins,
alpha(4)/alpha(9) integrins, alpha(6) integrins, and CD9 in the interaction of
the fertilin beta (ADAM2) disintegrin domain with the mouse egg membrane. Biol Reprod 66,1193–1202.
Zhu X, Bansal NP and Evans JP (2000) Identification of key functional amino
acids of the mouse fertilin beta (ADAM2) disintegrin loop for cell–cell
adhesion during fertilization. J Biol Chem 275,7677–7683.
Zuccotti M, Giorgi Rossi P, Fiorillo E, Garagna S, Forabosco A and Redi
CA (1998) Timing of gene expression and oolemma localization of
mouse alpha6 and beta1 integrin subunits during oogenesis. Dev Biol
200,27–34.
Submitted on February 2, 2005; resubmitted on May 18, 2005; accepted on
July 1, 2005